Dynamic feedback in a recommendation system

1. A computer-implemented method comprising:

selecting, based on a profile vector of a user, a plurality of seed media items from among a plurality of media items available for playback to the user within a computerized media recommendation system, wherein the profile vector comprises a normalized sum of media item vectors, the media item vectors being vector representations of media items previously consumed by the user, the vector representations produced from at least one of a word embedding process performed on the media items previously consumed by the user, a neural network process performed on the media items previously consumed by the user, or a collaborative filtering process performed on the media items previously consumed by the user, wherein a selection of the plurality of seed media items is based on first dot products between the profile vector and a vector representation of a first subset of the plurality of media items;

providing a representation of each seed media item of the plurality of seed media items to the user;

receiving a selection of a first seed media item from among the plurality of seed media items from the user;

automatically generating a pool of media items based upon the first seed media item, the pool comprising a subset, less than a whole, of the plurality of media items available within the computerized media recommendation system;

receiving a first feedback regarding a first media item from the user; and

modifying contents of the pool based upon the first feedback to generate a modified pool, wherein the modified pool includes at least one media item not included in the pool, excludes at least one media item included in the pool, or both.

2. The method of claim 1 , further comprising:

selecting a second media item from the pool based upon the first media item.

3. The method of claim 2 , wherein the first media item is represented in the computerized media recommendation system as a first vector, and the second media item is represented in the computerized media recommendation system as a second vector.

4. The method of claim 3 , wherein the second vector is closer to the first vector than any vector of any other media item in the pool.

5. The method of claim 3 ,

wherein the first feedback is a negative feedback, and

wherein the modifying the contents of the pool comprises removing the at least one media item included in the pool, the at least one media item included in the pool having a vector that is closer to the first vector than any vector of any other media item in the pool.

6. The method of claim 5 , wherein the negative feedback comprises one user interaction selected from the group consisting of: a dislike and a skip.

7. The method of claim 3 ,

wherein the first feedback is a positive feedback, and

wherein the modifying the contents of the pool comprises adding the at least one media item not included in the pool from among the plurality of media items available within the computerized media recommendation system.

8. The method of claim 7 , wherein the positive feedback comprises one user interaction selected from the group consisting of: a like, a play, and a full play.

9. The method of claim 1 , wherein the modifying occurs subsequent to the receiving the first feedback, subsequent to beginning playback of the first seed media item, prior to completing playback of a subsequent media item, and prior to receiving feedback from the user regarding the subsequent media item.

10. The method of claim 1 , further comprising:

beginning playback of a subsequent media item from the modified pool;

receiving a second feedback from the user regarding the subsequent media item; and

further modifying the modified pool to generate a second modified pool, based upon the second feedback.

11. The method of claim 1 , wherein the selection of the plurality of seed media items is further based on second dot products between the vector representation of the first subset of the plurality of media items and a vector representation of a second subset of the plurality of media items.

12. A computerized media recommendation system comprising:

a processor configured to:

select, based on a profile vector of a user, a plurality of seed media items from among a plurality of media items available for playback to the user within a computerized media recommendation system, wherein the profile vector comprises a normalized sum of media item vectors, the media item vectors being vector representations of media items previously consumed by the user, the vector representations produced from at least one of a word embedding process performed on the media items previously consumed by the user, a neural network process performed on the media items previously consumed by the user, or a collaborative filtering process performed on the media items previously consumed by the user, wherein a selection of the plurality of seed media items is based on first dot products between the profile vector and a vector representation of a first subset of the plurality of media items;

provide a representation of each seed media item of the plurality of seed media items to the user;

receive a selection of a first seed media item from among the plurality of seed media items from the user;

automatically generate a pool of media items based upon the first seed media item, the pool comprising a subset, less than a whole, of the plurality of media items available within the computerized media recommendation system;

receive a first feedback regarding a first media item from the user; and

modify contents of the pool based upon the first feedback to generate a modified pool, wherein the modified pool includes at least one media item not included in the pool, excludes at least one media item included in the pool, or both.

13. The system of claim 12 , the processor further configured to:

select a second media item from the pool based upon the first media item.

14. The system of claim 13 , wherein the first media item is represented in the computerized media recommendation system as a first vector, and the second media item is represented in the computerized media recommendation system as a second vector.

15. The system of claim 14 , wherein the second vector is closer to the first vector than any vector of any other media item in the pool.

16. The system of claim 14 ,

wherein the first feedback is a negative feedback, and

wherein the processor is configured to modify the contents of the pool by removing the at least one media item included in the pool, the at least one media item included in the pool having a vector that is closer to the first vector than any vector of any other media item in the pool.

17. The system of claim 16 , wherein the negative feedback comprises one user interaction selected from the group consisting of: a dislike and a skip.

18. The system of claim 14 ,

wherein the first feedback is a positive feedback, and

wherein the processor is configured to modify the contents of the pool by adding the at least one media item not included in the pool from among the plurality of media items available within the computerized media recommendation system.

19. The system of claim 18 , wherein the positive feedback comprises one user interaction selected from the group consisting of: a like, a play, and a full play.

20. The system of claim 12 , wherein the processor is further configured to modify the contents of the pool subsequent to receiving the first feedback, subsequent to beginning playback of the first seed media item, prior to completing playback of a subsequent media item, and prior to receiving feedback from the user regarding the subsequent media item.

21. The system of claim 12 , the processor further configured to:

begin playback of a subsequent media item from the modified pool;

receive a second feedback from the user regarding the subsequent media item; and

further modify the modified pool to generate a second modified pool, based upon the second feedback.

22. The system of claim 12 , wherein the selection of the plurality of seed media items is further based on second dot products between the vector representation of the first subset of the plurality of media items and a vector representation of a second subset of the plurality of media items.